Method for dynamic prior image constrained image reconstruction
A method for reconstructing a high quality image from undersampled image data is provided. The image reconstruction method is applicable to a number of different imaging modalities. Specifically, the present invention provides an image reconstruction method that incorporates an appropriate prior image into the image reconstruction process. Thus, one aspect of the present invention is to provide an image reconstruction method that requires less number of data samples to reconstruct an accurate reconstruction of a desired image than previous methods, such as, compressed sensing. Another aspect of the invention is to provide an image reconstruction method that produces a time series of desired images indicative of a higher temporal resolution than is ordinarily achievable with the imaging system. For example, cardiac phase images can be produced with high temporal resolution (e.g., 20 milliseconds) using a CT imaging system with a slow gantry rotation speed.
1. A method for reconstructing an image of a subject with an imaging system, the steps comprising:
a) acquiring, with the imaging system, a time series of image data of the subject;
b) selecting an estimate image of the subject;
c) reconstructing a prior image of the subject from the time series of image data acquired in step a);
d) producing a sparsified image of the subject using the prior image and the estimate image; and
e) reconstructing a desired image of the subject using the sparsified image, the estimate image, and the acquired time series of image data.
2. The method as recited in claim 1 in which steps b)-e) are repeated to produce a time series of desired images.
3. The method as recited in claim 2 in which the time series of image data is acquired with a first temporal resolution and the time series of desired images is indicative of a second temporal resolution that is higher than the first temporal resolution.
4. The method as recited in claim 1 further comprising:
f) acquiring a signal indicative of subject motion; and
g) producing a motion phase image data set from the acquired time series of image data.
5. The method as recited in claim 4 in which step g) includes:
selecting a gating window indicative of a time period in the acquired signal indicative of subject motion; and
selecting, from the acquired time series of image data, image data acquired during the gating window.
6. The method as recited in claim 5 in which step g) further includes:
dividing the selected gating window into a plurality of time windows; and
dividing the motion phase image data set into a corresponding plurality of motion subphase image data sets using the plurality of time windows.
7. The method as recited in claim 4 in which step b) includes reconstructing the estimate image from the motion phase image data set.
8. The method as recited in claim 4 in which the subject motion is physiological motion of the heart, the signal indicative of the subject motion is an electrocardiogram (ECG) signal, and the motion phase image data set is a cardiac phase image data set.
9. The method as recited in claim 1 in which step d) includes subtracting the prior image from the estimate image.
10. The method as recited in claim 1 in which step e) includes:
e)i) producing an objective function using the sparsified image, the estimate image, and the acquired time series of image data; and
e)ii) reconstructing the desired image by iteratively minimizing the objective function.
11. The method as recited in claim 10 in which step e)i) includes:
producing at least a first and second objective function term, wherein the at least first objective function term is produced by applying a sparsifying transform to the sparsified image and the at least second objective function term is produced by applying a sparsifying transform to the estimate image; and
adding the at least first and second objective function terms.
12. The method as recited in claim 11 in which step e)i) further includes calculating a norm of the at least first and second objective function terms and weighting the calculated norms using a regularization parameter.
13. The method as recited in claim 10 in which the objective function has at least one of the following forms:
α∥Ψ 1 ( I−I P )∥ p +(1−α)∥Ψ 2 I∥ p ;
α∥Ψ 1 ( I−I P )∥ p +(1−α)∥Ψ 2 I∥ p +λ∥X∥ 2 2 ; and
α∥Ψ 1 ( I−I P )∥ p +(1−α)∥Ψ 2 I∥ p +λ( X T DX );
wherein:
α is a regularization parameter;
Ψ 1 is a first sparsifying transform;
Ψ 2 is a second sparsifying transform;
I is the estimate image;
I P is the prior image;
∥ . . . ∥ p is an L p -norm;
λ is a Lagrange multiplier;
X is a difference matrix indicative of the difference between the acquired time series of image data and a forward projection of the estimate image; and
D is a system noise matrix.
14. The method as recited in claim 1 in which the imaging system is at least one of a magnetic resonance imaging (MRI) system, an x-ray computed tomography (CT) imaging system, a C-arm x-ray imaging system, a positron emission tomography (PET) imaging system, and a single photon emission computed tomography (SPECT) imaging system.
15. A method for reconstructing a time series of images of a subject with an imaging system, the steps comprising:
a) acquiring, with the imaging system at a first temporal resolution, a time series of image data of the subject;
b) selecting an estimate image of the subject;
c) reconstructing a prior image of the subject from the time series of image data acquired in step a);
d) producing a sparsified image of the subject using the prior image and the estimate image; and
e) reconstructing a time series of desired images of the subject using the sparsified image, the estimate image, and the acquired time series of image data, wherein the time series of desired images of the subject is indicative of a second temporal resolution that is higher than the first temporal resolution.
16. The method as recited in claim 15 further comprising:
f) acquiring a signal indicative of subject motion; and
g) producing a motion phase image data set from the acquired time series of image data.
17. The method as recited in claim 16 in which step g) includes:
selecting a gating window indicative of a time period in the acquired signal indicative of subject motion; and
selecting, from the acquired time series of image data, image data acquired during the gating window.
18. The method as recited in claim 17 in which step g) further includes:
dividing the selected gating window into a plurality of time windows; and
dividing the motion phase image data set into a corresponding plurality of motion subphase image data sets using the plurality of time windows.
19. The method as recited in claim 16 in which step b) includes reconstructing the estimate image from the motion phase image data set.
20. The method as recited in claim 16 in which the subject motion is physiological motion of the heart, the signal indicative of the subject motion is an electrocardiogram (ECG) signal, and the motion phase image data set is a cardiac phase image data set.